You are on page 1of 37

Management Publications Management

1-2007

Moral Judgment and Causal Attributions:


Consequences of Engaging in Earnings
Management
Steven E. Kaplan
Arizona State University

James C. McElroy
Iowa State University

Sue Ravenscroft
Iowa State University, sueraven@iastate.edu

Charles B. Shrader
Iowa State University, cshrader@iastate.edu

Follow this and additional works at: http://lib.dr.iastate.edu/management_pubs


Part of the Business Law, Public Responsibility, and Ethics Commons, and the Finance and
Financial Management Commons
The complete bibliographic information for this item can be found at http://lib.dr.iastate.edu/
management_pubs/8. For information on how to cite this item, please visit http://lib.dr.iastate.edu/
howtocite.html.

This Article is brought to you for free and open access by the Management at Iowa State University Digital Repository. It has been accepted for
inclusion in Management Publications by an authorized administrator of Iowa State University Digital Repository. For more information, please
contact digirep@iastate.edu.
Moral Judgment and Causal Attributions: Consequences of Engaging in
Earnings Management
Abstract
Recent, well-publicized accounting scandals have shown that the penalties outsiders impose on those found
culpable of earnings management can be severe. However, less is known about how colleagues within internal
labor markets respond when they believe fellow managers have managed earnings. Designers of responsibility
accounting systems need to understand the reputational costs managers impose on one another within
internal labor markets. In an experimental study, 159 evening MBA students were asked to assume the role of
a manager in a company and respond to a scenario in which another manager (the target manager) has the
opportunity to engage in earnings management. Participants provided causal attributions, assessed the
morality of the target manager, and indicated whether they would change their judgments about the target
managers reputation. The study manipulated three between-subjects factors: (1) whether the target manager
chose to engage in earnings management, (2) whether the companys budgetary control system was rigid or
flexible, and (3) whether the target managers work history was average or above average. We found that causal
attributions are affected more by the budgetary systems when the target did not manage earnings than when
the manager did. We also found that morality judgments were significantly associated with the target
managers behavior, but not with the budgetary system. In addition, participants judgments about the target
managers reputation were more strongly associated with morality judgments than with causal attributions.
We discuss implications of the role of reputation in management control systems design.

Keywords
Accounting, budgetary control system, casual attributions, earnings management, moral judgement, work
history

Disciplines
Business Law, Public Responsibility, and Ethics | Finance and Financial Management

Comments
This is a manuscript of an article from Journal of Business Ethics 74 (2007): 149, doi: 10.1007/
s10551-006-9226-y. Posted with permission. The final publication is available at Springer via
http://dx.doi.org/ 10.1007/s10551-006-9226-y.

This article is available at Iowa State University Digital Repository: http://lib.dr.iastate.edu/management_pubs/8


This is a manuscript of an article from Journal of Business Ethics 74 (2007): 149, doi: 10.1007/s10551-006-9226-y. Posted with permission. The
final publication is available at Springer via http://dx.doi.org/ 10.1007/s10551-006-9226-y.

MORAL JUDGMENT AND CAUSAL ATTRIBUTIONS: CONSEQUENCES OF

ENGAGING IN EARNINGS MANAGEMENT

Steven E. Kaplan
School of Accountancy and Information Management
Arizona State University
Tempe, AZ 85287-3606
Phone: 480 965 6498

Susan P. Ravenscroft
Department of Accounting
Iowa State University
Ames, IA 50011
Phone 515 294 3574
FAX 515 294 3525

James C. McElroy
Department of Management
Iowa State University
Ames, IA 50011
Phone 515 294 4277
FAX 515 294 7112

C. Bradley Shrader
Department of Management
Iowa State University
Ames, IA 50011
Phone 515 294 3050
FAX 515 294 7112
cshrader@iastate.edu

August 13, 2004

Submitted to the Journal of Business Ethics: Accounting and Finance Section


MORAL JUDGMENT AND CAUSAL ATTRIBUTIONS: CONSEQUENCES OF

ENGAGING IN EARNINGS MANAGEMENT

ABSTRACT

In an experimental study, one hundred fifty-nine evening MBA students were asked to

assume the role of a manager in a company in which another manager has the opportunity to

engage in earnings management. In response to the scenario, participants provided causal

attributions, assessed the morality of the target, and indicated whether they would change their

judgments about the targets reputation. The study manipulated three between-subjects factors:

1) whether the target manager chose to engage in earnings management, 2) whether the

companys budgetary control system was rigid or flexible, and 3) whether the target managers

work history was average or above average. The results indicated an interactive effect such that

causal attributions differ more across different budgetary systems when the hypothetical manager

did not manage earnings than when the manager did. The results also indicated that morality

judgments were significantly associated with the hypothetical managers behavior, but not with

budgetary system. In addition, the judgments subjects provided about the managers reputation

were found to be more strongly associated with morality judgments than with causal attributions.

We discuss implications of the role of reputation in management control systems design.

Key words: Causal Attributions, Earnings Management, Moral Judgment

1
MORAL JUDGMENT AND CAUSAL ATTRIBUTIONS: CONSEQUENCES OF

ENGAGING IN EARNINGS MANAGEMENT

Healy and Wahlen define earnings management as using judgment in reporting financial

results "and in structuring transactions to either mislead some stakeholders about the underlying

economic performance of the company or to influence contractual outcomes that depend on

reported accounting numbers" (1999, p. 368). More recently, Davidson, Jiraporn, Kim and

Nemec (2004, p. 267) have conceptualized earnings management as a form of impression

management whereby flexible accounting principles are used to influence reported earnings,

thereby causing reported income to be larger or smaller than it might otherwise be. Opportunities

for earnings management exist because managers typically have the ability to use judgment in

shaping financial reports (Jensen 2001), both in setting earnings targets and in reporting actual

results. Healy and Wahlen (1999) conclude that earnings management is motivated by both

capital market factors and/or management compensation considerations.

As an example of the latter motivation, Guidry et al. (1999) use business-unit level data

to demonstrate that earnings management occurs when managers attempt to maximize or

increase their bonuses. In their field study, Guidry et al. (1999) found that business-unit

managers in the bonus range appeared to manage earnings upward through the use of

discretionary accruals, as compared to business-unit managers who were not in the bonus range.

This evidence demonstrates that, on average, managers act as if the benefits from engaging in

earnings management outweigh potential negative consequences whether internal or external to

the firm. Simply put, a variety of inducements exist that may motivate managers to manipulate

both earnings targets and reported results (Jensen, 2001).

2
While evidence indicates that some managers engage in earnings management in order to

receive performance-based bonuses (Guidry et al. 1999; Healy, 1985), others refrain because

they consider earnings management activities to be unethical (Merchant and Rockness, 1994;

Elias, 2004). Accounting researchers have not explored the consequences of earnings

management, i.e. whether managers who are discovered to have engaged in earnings

management to enhance their compensation suffer any negative effects on their reputations

within their firms.

Our work is motivated in part by Sprinkle (2001, p. 2), who suggests that management

accounting researchers should investigate the extent to which social norms, individuals

preferences for non-pecuniary factors such as honesty and fairness, and firms information

systems interact with more formal managerial accounting systems. The purpose of this study is

to examine the reputational consequences when managers engage in earnings management.

Specifically, the study proposes that observers morality assessments and causal attributions

about a target manager who has an earnings management opportunity will be shaped by both the

target managers behavior and the organizations budgetary control style. Further, it is suggested

here that effects on the targets reputation will be significantly associated with morality

assessments about the target manager. We believe that it is important for designers of

responsibility accounting systems to understand these reputational costs as they design

management control systems and formal information systems within firms.

In particular, in this study we ask subjects to judge the morality of earnings management

decisions in differing budgetary and performance contexts to determine what causal attributions

are given and how morality judgments and attributions affect others impressions about the

manager who engages in such behavior. Earnings management has been an area of longstanding

3
concern in the field of accounting, but relatively little research has addressed morality judgments

of the practice from a management perspective. Though research directly addressing the issue is

lacking, it is generally felt that managers tend to be morally sensitive to differing work contexts

(Brower and Shrader, 2000) and ethical climates (Forte, 2004) in making decisions. Research

addressing specific moral judgments potentially help us understand the nature of corporate

leadership in organizations (Kelly, 2004). To this end, we seek to empirically explore how

various specific situations interact with moral judgments of earnings management.

BACKGROUND AND HYPOTHESES

Earnings Management

Merchant and Rockness contend that earnings management is probably the most

important ethical issue facing the accounting profession," (1994, p. 92). Given the recent

Enron environment, this is even truer today. Merchant and Rockness (1994) provide initial

evidence on how professionals assess the ethics of earnings management activities. The

respondents, selected from two corporations and one chapter of the Institute of Internal Auditors,

were general managers, corporate staff, or operating unit controllers. They were asked to answer

several questions based on a series of brief scenarios, each of which described an earnings

management activity undertaken by a hypothetical profit center manager. Participants were asked

to evaluate the acceptability of each activity. General managers rated the activities as more

unethical than the corporate staff or unit controllers did. This finding is consistent with earnings

management having greater negative consequences for line managers than for staff. Presumably,

non-managers such as corporate staff employees or internal auditors are not competing against

other managers for compensation or promotion. The responses from the corporate employees

differed significantly, causing Merchant and Rockness to speculate that respondents from

4
Corporation A may have rated the activities more unethically than did respondents from

Corporation B because of one or more difficult-to-observe control factors, such as tone at the

top (1994, p. 91). Additional research by Kaplan (2001a) showed that reactions to earnings

management varied by the observers role. Those assigned the role of stakeholder saw any

earnings management attempt as unethical, while those assigned the role of another manager in

the company took a differentiated view of earnings management. In this case, subjects assigned

the role of managers viewed accounting-related earnings management as more unethical than

operations-related earnings management.

To a degree this study extends the work of Kaplan (2001a), by further examining the

reactions of other managers within a firm to accounting-related attempts at earnings

management. Rather than simply looking at the ethical judgments associated with accounting-

related earnings management, this study seeks to ascertain the additional roles played by the

target managers record of past performance as well the organizations budgetary control system.

Attribution Theory

Attribution theory offers a framework for understanding the processes by which people

explain, interpret, and respond to the behavior of others (Kelley and Michela, 1980; Weiner,

1985a). Attribution theorists are concerned with perceptions of causality, or the post-hoc reasons

individuals use to explain events (Heider, 1958). In essence, attribution processes involve post

hoc reasoning whereby the causes of a behavior or an event are inferred from the observation of

the behavior or event (Mowday, 1983). Such post-hoc sense-making is important because

individuals use it both to identify the causes of events and to assign personal qualities or

attributes to others engaging in behavior (Lord and Smith, 1983). Specifically, causal

attributions affect and reflect both how we feel about events and people and our expectations of

5
those people (Weiner, 1985b). Thus, the particular causes that individuals attribute to a given

behavior or event are closely related to more generalized evaluations (e.g., correspondent

inferences) of those individuals engaging in behavior (Crant and Bateman, 1993).

The causal explanations people use to explain events vary on a number of dimensions

(Weiner, 1985b), but for the purposes of this study the locus dimension is most relevant. That is,

human behavior can be explained along a continuum in which the end-points are dispositional

causes (e.g., internal to the individual engaging in behavior) and situational causes (e.g., external

to the individual engaging in the behavior) (Kelley and Michela, 1980). The former is primarily

concerned with an individuals assessment of the extent to which an event is caused by the

character, disposition or intent of the person engaging in the event while the latter is due to

situational factors.

The application of attribution theory to organizational settings was pioneered by the work

of Green and Mitchell (1979), among others, and has more recently been applied to examine

auditor judgments (e.g., Kaplan and Reckers, 1985, 1991, 1993) and jurors evaluations (Lowe,

Reckers, and Whitecotton, 2002). Attribution theory is most applicable when causality is

uncertain. Understanding why a manager did or did not engage in earnings management involves

uncertainty because frequently one cannot inquire too closely about or definitively ascertain that

such behavior has in fact occurred.

In this study we explore the attributional processes occurring when an earnings

management opportunity exists. Previous earnings management research is extended in three

ways. First, the current study examines both the causal attributions and a morality-based

correspondent inference (e.g., an overall assessment formed about the target managers morality)

individuals form in response to an earnings management opportunity by a target manager.

6
Whereas causal attributions have been examined extensively in auditor judgments and earnings

forecasts (Kaplan and Reckers, 1985, 1991, 1993; Lowe et al., 2002; Baginski et al., 2004), the

formation and implications of correspondent inferences has received less attention (Wong-on-

Wing et al., 1989; Reckers and Wong-on-Wing, 1991). We test an alternative hypothesis based

on Reeder and Spores (1983), who have specified a particular type of ethics-based correspondent

inference to denote an individuals ethics or lack thereof. We provide further evidence on the

antecedents and consequences of this inference.

Second, the current study examines whether an important component of the

organizational environment, the budgetary control style (Van der Stede, 2000), is associated

either directly or indirectly with the formation of causal attributions and/or ethically related

correspondent inferences. As discussed below, the background and context in which behavior

occurs are predicted to influence and shape causal attributions and correspondent inferences. We

contend that any informal discipline in internal labor markets would be contingent upon the

extent to which the occurrence of an unethical event (e.g., engaging in earnings management) is

attributed internally, that is, to the disposition or character of the manager. That is, negative

outcomes and responses directed towards a target manager are expected only when other

managers make internal attributions and inferences about the target manager rather than external

attributions that serve to excuse or exculpate the target manager. In this regard, we would not

expect that a target managers reputation would be damaged if engaging in earnings management

is attributed primarily to strong environmental pressures, such as those present in certain

budgetary control systems.

Third, we examine the relationship between causal attributions and ethically-based

correspondent inferences, respectively, and subsequent reputationally-related judgments. While

7
prior research (Kaplan, 2001a, 2001b) has examined ethics-based correspondent inferences as a

dependent measure, we contend that such judgments drive subsequent reputationally-related

judgments to a greater extent than causal attributions do.

Antecedents to Causal Attributions and Correspondent Inferences

Kelley (1972) introduced both the discounting and augmentation principles to the

attribution literature. Under the discounting principle, engaging in behavior that is expected in

the situation lessens the tendency to make internal attributions and correspondent inferences to

the individual. Alternatively, under the augmentation principle, engaging in contraindicated

behavior (Michela and Kelley, 1980, p. 470) will strengthen the tendency to make internal

attributions and correspondent inferences to the individual. We consider the application of these

principles to observing a target managers decision regarding earnings management.

The situational variable that could either lead to either discounting or augmentation that

we examine is the budgetary control system. While budgetary controls have been defined in

various ways, the issue has generated strong and continuing interest among management

accounting researchers (Hopwood, 1972; Merchant, 1998; Van der Stede, 2000). Budgetary

control styles fall along a continuum ranging from rigid to flexible. A rigid budgetary control

style provides a strong incentive for meeting accounting-based budgets (Merchant, 1998) and is

one in which managers are evaluated primarily on whether or not they achieved their short-term

accounting-based budget. A flexible budgetary control style places less reliance upon

accounting-based budget information and considers such information in context with other

information in a longer-term time frame.

Assuming that budgetary control style leads to discounting or augmentation, we propose

the following relationships. Ceteris paribus, when engaging in earnings management allows

8
managers to meet their financial targets and when managers work under a rigid budgetary control

style, observers would expect managers to manage earnings. Hence, the observers attributions

would reflect discounting, i.e. less emphasis on internal or dispositional explanations. Similarly,

when managers refrain from earnings management in a flexible budget regime, they are behaving

as expected and again observers attributions would reflect discounting. Alternatively, when

budgetary control styles are rigid and the target manager faces pressure to manage earnings, his

decision not to do so will lead to attributions that reflect augmentation, i.e. a greater emphasis on

disposition. In the same vein, a target manager who chooses earnings management even though

he faces a flexible budgetary control system is doing the unexpected and internal attributions

made about him will be augmented.

Hypothesis 1 (a): Earnings management and budgetary control style will interact such
that under a flexible control setting managers causal attributions will be more internally
oriented when the actor engages in earnings management. Alternatively, under a rigid
control setting managers causal attributions will be more internally oriented when the
manager does not engage in earnings management.

We propose that observers will make morality judgments about the target manager that

reflect their attributions. When the target behaves as the budgetary control style would lead him

to, then the morality judgment of him is expected to be relatively neutral. But if he acts contrary

to expectations then the judgments will be more extreme. In the case of managing earnings in a

flexible budgetary setting he will be seen as unethical (e.g., less moral), in contrast to a rigid

budgetary setting where a more neutral assessment is expected. In the case of refraining from

managing earnings when the budgetary controls are rigid he will be seen as highly moral, in

contrast to a flexible budgetary setting where a more neutral assessment is expected. As

described, morality judgments about the manager will be more favorable under a rigid budgetary

9
setting rather than a flexible budgetary setting. This argument leads to a related hypothesis

regarding morality assessments.

Hypothesis 1 (b): Morality judgments about the target manager will be more favorable
when budgetary control style is rigid rather than flexible.

However, a somewhat different pattern of attribution is predicted by other theorists

(Reeder and Spores, 1983; Erickson and Krull, 1999). They suggest that because negative

information weighs heavily in ones overall impression of a person, a single immoral behavior,

such as engaging in earnings management, may be enough to sour ones overall evaluation of a

person (Reeder and Spores, 1983). This heightened role of negative information has been found

by auditing researchers (Butt and Campbell, 1989; Kida, 1984; Trotman and Sng, 1989).

Moreover, Reeder and Spores (1983), contend that when people observe unethical behavior they

tend to focus on that behavior and to generalize from that observation. Under their model

observers tend to believe that if someone behaves unethically in one situation, they are likely to

do so in other situations and in other ways (cf. also Sanderson and Darley, 2002). Thus when the

target engages in immoral behavior, an observer infers that the actor is immoral, regardless of the

situational demands surrounding the behavior (Reeder and Spores, 1983) and holds that person

accountable for his/her actions.

However, when ethical behavior is observed (e.g., not engaging in earnings

management), the ethical behavior may reflect either the target person prudence (i.e., he would

prefer to engage in earnings management but there is no immediate payoff under flexible

budgetary systems so he reports earnings accurately) or it may reflect good character. More

recently, Erickson and Krull (1999) have also argued that morality judgments are more extreme

than causal attributions, by suggesting that observers infer a dispositional or character trait from

10
behavior and, depending on circumstances may look for causal explanations only after forming a

morality judgment.

To test these alternative explanations, we posit the following set of competing hypotheses

on the role of earnings management and budgetary control systems on causal attributions and

morality judgments.

Hypothesis 2 (a): Earnings management and budgetary control style will interact such
that managers causal attributions differ across budgetary control systems when the actor
does not engage in earnings management (e.g., attributions will be more internally
oriented under a rigid control setting than flexible control setting) but will not differ
across budgetary control systems when the manager does engage in earnings
management.

Hypothesis 2 (b): Earnings management and budgetary control systems will interact
such that morality judgments about the target will differ across budgetary control systems
when the actor does not engage in earnings management (e.g., attributions will be more
internally oriented under a rigid control setting than flexible control setting) but will not
differ across budgetary control systems when the actor engages in earnings management.

Reputation Effects

Because our purpose is to examine possible reputational consequences of earnings

management, it is appropriate to consider how attributions relate to a range of judgments

and evaluations that managers may rely upon in forming opinions about their colleagues.

If, as Erickson and Krull (1999) suggest, morality judgments differ from causal

attributions, one of the differences may be that the process of generalization from moral

judgments is stronger (broader in its range and more immediate) than generalizations

from causal attributions. Consequently, we hypothesize that when observers assess the

reputation of a target manager, their reputational judgments will be more strongly

associated with morality assessments than with causal attributions.

Hypothesis 3: Reputation effects will be more strongly associated with morality


judgments than with causal attributions.

11
METHOD

Overview and Task

The subjects were presented with a scenario describing an earnings management

opportunity for a target manager, Mr. Jones, a divisional manager for a public company.

Participants were instructed to assume the role of another manager working for this same

company. The earnings management opportunity involved the possibility that Mr. Jones could

defer receipt and related expense accrual of a bill for $220,000, which is part of an ongoing

consulting engagement. Participants received a single, randomly assigned case. In response to

the scenario, participants provided a series of responses described below. To complete the

questionnaire, participants responded to manipulation checks and several questions about their

background.

Independent Variables

The study contained three between-subjects independent variables, as follows: behavior

of the target manager, the nature of the organizations budgetary control system, and the target

managers work history.

Earnings Management Behavior

The target managers behavior was operationalized as a binary variable with two levels.

In the earnings management condition, the scenario indicated that the target manager engaged in

the earnings management activity. Under this manipulation, the case read, in part:

For the most recent period, the division reached targeted net income.
You learn that divisional expenses did not include the costs from an
ongoing consulting engagement. Mr. Jones called the engagement
partner of a consulting firm that was doing some work for the division
and asked that the firm not send an invoice until next year. The partner
agreed. Mr. Jones knew this action was questionable but that it would

12
allow the division to reach targeted net income. Divisional net income
would have been below targeted net income had the expenses from the
consulting engagement been included as part of this years divisional
expenses.

In the non-earnings management condition, the scenario indicated that the target manager did not

engage in the earnings management activity. Subjects in this condition read a case that said, in

part:

For the most recent period, the division did not reach targeted net
income. You learn, however, that divisional expenses included the costs
from an ongoing consulting engagement. Mr. Jones considered calling
the engagement partner of a consulting firm that was doing some work for
the division and asking that the firm not send an invoice until next year.
While Mr. Jones was sure that the partner would agree, he concluded that
such an action would be wrong. Divisional net income would have been
above the targeted net income had the expenses from the consulting
engagement been excluded from this years divisional expenses.

Budgetary Control Style

The nature of the organizations budgetary control style was manipulated to alter the

situational demands confronting the hypothetical manager. Under the rigid style, the case

described a budgetary control system with a heavy emphasis on meeting short-term targets.

Specifically, the case read, in part:

As a manager you have found the companys reward structure to have a


strong orientation towards short-term performance. As an example, a
budgeted annual income target is set for each division and it is important
for division managers to meet the target. Division managers achieving
budgeted annual income targets receive favorable evaluations and a
substantial bonus. However, division managers who do not meet budgeted
incomes receive unfavorable evaluations and do not get the bonus. The
extent to which the division is progressing towards long term strategic
goals has almost no impact on the managers performance evaluation.

Under the flexible style, the case described a budgetary control system with less emphasis

placed on meeting short-term targets. Specifically, the case read, in part:

13
As a manager you have found the companys reward structure to have a
strong orientation towards long-term performance. To enhance
coordination and communication, a budgeted annual income target for
each division is established. However, within the company whether the
budgeted annual income targets are achieved is not important. For
example, the managers bonus is not tied to achieving this budget and
whether the budget is achieved has little influence on the division
managers overall performance evaluation. Instead, division managers
are evaluated primarily with respect to measures more directly tied
towards meeting long-term strategic objectives. Also, bonuses are tied to
measures of long-term strategic goals.

Target Managers Work History

The target managers work history was manipulated at two levels. Under the average

work history level, the case read, in part, He has a reputation for being hard-working and

knowledgeable, and his division and his evaluations have been average. Under the above

average work history level, the case read, in part, He has a reputation for being hard-

working and knowledgeable, and his division and his evaluations have been very

favorable.

Dependent Variables

The study includes three groups of dependent variables: (1) causal attributions, (2)

morality judgments, and (3) four other managerially related measures of reputation.

Attribution Measure

Participants were asked to indicate the degree to which (1) factors related to Mr. Joness

disposition or character (e.g., internal scale), and (2) factors related to the situation and the

organization (e.g., external scale) contributed to Mr. Joness action. Two separate nine-point

scales were presented, anchored by very unlikely cause (1) and very likely cause (9). A net

attribution score for each participant was computed by subtracting the rating on the external scale

from the rating on the dispositional scale. The net attribution score has a range of -8 (situational

14
causality) to 8 (dispositional causality). This net score indicates the relative importance of factors

related to the managers internal traits vs. situational factors. A positive score indicates that the

participant attributed the managers action more to dispositional (internal) causes than to

situational (external) causes. A negative score indicates that the managers action was attributed

more to the situation than to the managers character. A score of zero indicates that the two

factors contributed equally to the managers behavior. This method of netting attribution

measures is common (Elig and Frieze, 1979) and has been used previously to measure auditors

net attribution scores (Kaplan and Reckers, 1985; Wong-on-Wing et al., 1988).

Morality Measure

The second measure is participants assessment (or judgment) of the morality of the

hypothetical manager. Participants were asked, Based on the available information, how would

you judge the morality of Mr. Jones? The end-points on a nine-point scale were very immoral

(1) and very moral (9).

Other Managerial Reputation Effects

The study also examines other managerially relevant judgments. These include how

knowledgeable Mr. Jones is, how hard-working Mr. Jones is, Mr. Jones decision making ability,

and how willing the subject would be to work with Mr. Jones.

As part of the background information describing Mr. Jones, he was characterized as

having a reputation for being hard-working and knowledgeable. On two separate scales

participants were asked to indicate the extent to which Mr. Jones actions changed their

impressions of Mr. Jones in terms of his reputation for being (1) hard working, and (2)

knowledgeable. Each nine-point scale was anchored by Action greatly diminishes reputation

(1) and Action greatly strengthens reputation (9). Participants were also asked, Based on the

15
available information, how would you assess Mr. Joness decision-making ability during the

current year? The end-points on a nine-point scale were substantially below expectations (1)

and substantially above expectations (9). The mid-point (5) was labeled at expectations.

Finally, participants were asked, Based on the available information, how willing would you be

to work with Mr. Jones on an important company project? The end-points on a nine-point scale

were extremely unwilling (1) and extremely willing (9).

Subjects

Evening MBA students at a major metropolitan state university were used as participants

for the study. The students were enrolled in a managerial accounting course, which is taken in

the second year. Evening MBA students typically are older than full-time students and have

substantial work experience. One hundred sixty two questionnaires were distributed and 159

questionnaires were completed. The remaining three had missing data and were excluded from

the analysis. The majority of participants was male (65 percent) and had been involved in

preparing a budget or providing information for others preparing a budget (63 percent). In

addition, the mean age and professional work experience among participants were approximately

32 years and 10 years, respectively.

RESULTS

Manipulation Checks

After reading the case, participants answered manipulation checks to determine whether

they were aware of the budgetary control system and work history of the hypothetical manager.

Regarding the budgetary control style faced by the manager, participants were asked two

questions. The first question asked, How would you assess the culture of the company? on a

nine-point scale anchored by long-term performance oriented (1) to short-term performance

16
oriented (9). The second question asked, How important is it for division managers to achieve

budgeted annual income targets? on a nine-point scale anchored by not important at all (1) to

very important (9). Mean responses to these two questions among participants in the rigid

condition were 8.3 and 8.0, respectively. Mean responses to these two questions among

participants in the flexible condition were 3.4 and 4.2, respectively. On each of the two

questions, the differences between the two groups were significant (p<.01). Regarding the work

history of the hypothetical manager, participants were asked, Prior to the current year, how

would you assess the managers work history: The nine-point scale was anchored by very

below average (1) to very above average (9). Mean responses from the average and above

average work history treatment levels were 4.8 and 6.7, respectively. The two groups were

statistically different from one another (p<.01). These responses indicate that participants

attended to the manipulations of budgetary control system and target managers work history.

Hypothesis Testing

Table 1 presents the descriptive statistics and a correlation matrix for the dependent

variables used in this study. Hypotheses 1(a and b) and 2(a and b) were tested using a 2x2x2

between-subjects factorial design. The three independent variables of target behavior (earning

management versus no earnings management), the organizations budgetary control style (rigid

versus flexible) and the targets work history (average versus above average) were analyzed

using the general linear analysis of variance model, which accommodates unequal cell sizes.

Hypothesis 3 was tested using regression analysis.

------------------------------------
Insert Table 1 about here
------------------------------------

17
Table 2 reports the ANOVA results for the effects of earnings management, budgetary

control style and work history on net attributions and morality judgments. Significant main

effects were found for both earnings management (F=20.54, p<.001) and work history (F=4.65;

p<.05) on net attributions. Subjects turned to more internal/dispositional attributions when the

target manager did not engage in earnings management (Mean=1.54) than when the manager did

in engage in earnings management (Mean=-.58; p<.001). As shown in Table 2, earnings

management explained 12 percent of the variance in net attributions. Moreover, the behavior of

the target manager was attributed more to internal factors when the manager was described as an

average performer (Mean=.97) than when he was described as an above average performer

(Mean=.01; p<.05). In this case, work history explained 3 percent of the variance in net

attributions. However, it is the finding of a significant interaction effect (explaining 4 percent of

the variance) between the earnings management and the budgetary control style that is most

relevant to this study. Recall, that both Hypothesis 1(a) and Hypothesis 2(a) predict a significant

interaction between earnings management and budgetary control style. However, each predicts a

different interaction pattern. Figure 1 shows the nature of the observed interaction. The

observed pattern conforms to the pattern predicted by hypothesis 2(a). That is, net attributions

were not affected by budgetary control style when earnings management occurred but were

affected by budgetary control style when earnings management did not occur. The strongest

internal attributions were assigned to the actor who did not engage in earnings management in

the rigid budgetary control system (Mean = 2.60).

---------------------------------
Insert Figure 1 about here
---------------------------------

18
Table 2 reveals a strong, significant effect for earnings management on morality

assessments (F=251.95; p<.001), explaining 63 percent of the variance. The absence of a

significant main effect for budgetary control style fails to support Hypothesis 1(b). Further the

absence of a significant interaction effect for target behavior and the budgetary control style fails

to support Hypothesis 2(b). The finding of a significant interaction between earnings

management and work history, while not directly related to Hypothesis 1(b) or 2(b) is interesting,

nonetheless. Above average performers were judged more moral when they engaged in earnings

management than were average performers (Mabove avg.=4.21, Mavg=3.49), but slightly less moral

when they did not engage in earnings management (Mabove avg.=7.23, Mavg.=7.39). This

interaction effect explained 3 percent of the variance in morality assessments.

---------------------------------------
Insert Table 2 about here
--------------------------------------

Hypothesis 3 addresses the extent to which other managerially-relevant judgments are

associated with either causal attributions and/or morality assessments. These include assessments

of how much the reputation of the target has changed in terms of how hard-working and

knowledgeable the target is, his decision-making ability, as well as changes in the observers

willingness to work with the target. Regression analysis was used to determine the association

between causal attribution and morality assessments on each of the four dependent variables.

Table 3 shows the results of regressing the subjects net attributions and morality

assessments on the four reputation perceptions of the target manager: diligence (i.e. willingness

to work hard), knowledge, decision making capability, and their willingness to work with the

target manager. As shown in Table 3, morality assessments have a much greater and more

19
pervasive affect on reputational judgments about the target manager than do net attributions,

findings that support Hypothesis 3.

-----------------------------------

Insert Table 3 about here

-----------------------------------

DISCUSSION

While prior research (Healy, 1985; Guidry et al., 1999) has provided important insights

indicating that some managers opportunistically manage earnings, little attention has been given

to the potential internal negative consequences associated with engaging in such behavior. The

current paper provides evidence on these potential negative consequences in terms of causal

attributions observers make to explain such behavior, the morality judgments observers make,

and reputational consequences, including perceptions of their work ethic, knowledge, decision

making abilities and the future willingness of colleagues to work with them.

Limitations

Before discussing the results of the study, several limitations related to the use of an

experimental approach should be noted. As part of an experimental approach, participants

responded to a scenario about a target managers opportunity to engage in earnings management.

This approach has previously been used in studies to measure ethically-related judgments (Becker

and Fritzche, 1987; Flory et al., 1992; Singer and Singer, 1997), attributions, and performance-

related judgments (Kaplan and Reckers, 1985). The strengths of this approach include

experimental control and the ability to manipulate key variables. A concern, however, with this

approach is that it is not possible for the stimulus material to contain all relevant information that

would be available in a real-world setting. For example, although amounts were given for the

20
potential earnings management activity as well as the size of the company, other financial

statement information as well other non-financial information was not provided. Potentially, the

availability of such additional information might have influenced the judgments examined in this

study.

Second, the participants in the study were evening MBA students asked to assume the

role of a company manager. While the majority of these students had substantial professional

work experience and had been involved with the budgeting process, caution is warranted in

extrapolating these findings to the work environment. Third, the dependent measures used in the

study were almost exclusively one-item scales. The use of such measures leaves open questions

of the validity and reliability of these measures.

Results

The results of the study support Reeder and Spores (1983) notion (Hypothesis 2a) that

causal attributions would differ more across budgetary systems when a target manager had not

engaged in earnings management than when he had. This is based on the notion that immoral

behavior reflects immoral dispositions, while moral behavior may sometimes indicate moral

character but could also indicate prudent or strategic behavior by an immoral person. The data

provided support for the hypothesized relationship.

When the target manager behaved ethically (i.e., did not engage in earnings management)

the mean net attribution measure was positive, indicating that internal factors were perceived as

having greater influence on the target managers actions than the circumstances. This finding is

consistent with the actor-observer attributional effect reported by Jones and Nisbett (1972),

which showed that observers of a target person (actor) naturally turn to the actor per se as the

cause of an event. However, when the target manager engaged in earnings management the

21
mean net attribution measure was slightly negative, indicating that organizational factors were

perceived as having a slightly greater influence on the targets actions than did the character of

the target manager. In other words, the short term nature of the budget focus faced by the target

manager may have served as a potential (situational) explanation for the engagement in earnings

management. Given that most of the subjects in this study had experience with a budgetary

process, they may have empathized with the target manager who engages in earnings

management. Empathy for the situation faced by an actor has been shown to offset the actor-

observer effect (Regan and Totten, 1975), a finding similar to an effect found by Tan and Lipe

(1997), that MBA subjects were not likely to judge a targets actions harshly, even when the

outcomes of the targets decisions were negative.

The results supported a strong main effect of behavior on morality judgments, but neither

the predicted main effect for budgetary control style (Hypothesis 1b) nor the predicted

interaction between budgetary control style and behavior (Hypothesis 2b) was significant at

traditional levels. Thus, no support is found for either Hypotheses 1b or 2b. Respondents judged

target managers who did not engage in earnings management as being more moral, regardless of

the situational (budgetary control) constraints. The lack of a significant interaction effect may

reflect differences in the ethicality of events considered by the two studies. In their second

experiment Reeder and Spores used an incident where the subject sees someone drop a twenty-

dollar bill and chooses either to keep it or to call out to the person who dropped the money. Their

second incident involved stealing from a charity box set up by the cash register in a pizzeria.

These incidents may be less morally ambiguous than earnings management is, particularly in this

age of Enron. In this regard, Bruns and Merchant (1990) contend that there is a lack of

agreement among managers regarding the acceptability of earnings management activities.

22
While Merchant and Rockness (1994) showed a lack of consensus among professionals about the

morality of earnings management, the societal context surrounding earnings management today

appears to have resolved any such ambiguity.

The third hypothesis is based on the distinction Hamilton (1980) notes between ascribing

causation and assigning moral responsibility. We hypothesized that when the participants formed

a variety of reputation-related judgments about the target manager, these would be more closely

associated with their morality judgments than with their causal explanations (internal

attributions). Consistent with hypothesis 3, the results from each of the four reputation-related

judgments indicated a stronger association with morality judgments than with net attribution

judgments. These results support the contention by Erickson and Krull (1999) that morality

judgments are more extreme than causal attributions. Our results are also consistent with the

findings of Reeder and Spores (1983), who found that subjects are likely to generalize from their

morality judgments. That is, if a person behaves immorally in one context, observers infer that

the person will behave immorally in other ways.

Implications

Reputational consequences of ethical actions have not been widely studied in accounting.

These findings are important, however, because they indicate that managers engaging in earnings

management may face negative reputational costs from other managers, which in turn, may serve

as a disincentive towards engaging in earnings management. Several implications stem from this

research. First, to the degree that managers who engage in earnings management go unpunished,

or are perhaps even rewarded, the tendency to continue to engage in such practices is reinforced.

As the support for Hypothesis 2a suggests, observers tend not to blame an individual for earnings

management when the budgetary control system supports such behavior. Blaming the target for

23
behaving as rewarded results in what Kerr (1995) calls the folly of rewarding A while hoping for

B. That is, we would hope that managers would not engage in such behavior, but the system

rewards them for doing so.

Second, research has shown that failing to discipline has consequences not only for the

person who engages in suspect behavior, but also for observers of such behavior (Trevino, 1992).

Trevinos research shows that others within organizations expect rule violators to be punished in

order to maintain the social order and the punishment or failure to punish can be expected to

influence observers subsequent behaviors. In short, punishment or the failure to punish sends

very real messages throughout the organization.

The ethical reputation a manager has among colleagues can play an informal, but

significant, role in internal disciplining of managers by other managers. Research by Hollinger

and Clark (1983) on employee deviance demonstrated that the effect of informal sanctions was 2

times greater than that of formal sanctions. The results of this study offer some evidence on

what shapes the way workers view the morality of their colleagues. For example, we find

evidence that among participants there is a tendency to generalize from a morality judgment to

other dimensions that relate to competence, such as being hard working, knowledgeable, and

able to make decisions. Informal assessments managers make regarding their colleagues, based

at least in part on ethical concerns, can play a real but not well-measured or understood aspect of

management control systems. We have provided evidence that social costs are imposed on

managers when they do not behave in goal congruent ways. Cohen and Prusak (2001) reinforce

the importance of establishing social capital, i.e. goodwill and trust among employees within

firms. These authors maintain that high social capital results in better financial results. The loss

of trust that can be engendered by earnings management would be a loss of such social capital,

24
and is an issue that accounting researchers could examine more fully. Noreen (1988, p. 367)

noted that a cost of unethical behavior is a decline in ethical norms, which leads to an increase in

unethical behavior.

Given the paucity of research on the topic of informal, ethically-related aspects of control

systems, additional research should be encouraged. For example, further research could explore

whether the negative consequences vary across different kinds or magnitudes of earnings

management activities. In this regard, the results of the current study suggest that in spite of its

limitations, an experimental approach is a viable and appropriate method. Perhaps, within an

experimental setting the influence of ambiguity could be examined. For example, a future study

could manipulate the degree to which organizations include various types of earnings

management activities in following Section 406 of the Sarbanes-Oxley Act specifying the

inclusion of understandable disclosure in reports and documents filed with the SEC (Lander,

2004). Another possible avenue of further research is to look at how informal discipline within

internal labor markets can lead to changes or reductions in more formal monitoring procedures.

25
REFERENCES

Baginski, S. P., Hassell, J. M. and M. D. Kimbrough: 2004, Why Do Managers Explain Their Earnings
Forecasts? Journal of Accounting Research 42 (1), 1-28.

Becker, H., and D. J. Fritzsche: 1987, A Comparison of the Ethical Behavior of American, French and
German Managers, Columbia Journal of World Business 22, 87-95.

Butt, J. L., and T. L. Campbell: 1989, The Effects of Information Order and Hypothesis-Testing
Strategies on Auditors Judgments, Accounting, Organizations and Society 14, 471-479.

Brower, H. H. and C. B. Shrader: 2000, Moral Reasoning and Ethical Climate: Not-for-Profit vs. For-
Profit Boards of Directors, Journal of Business Ethics 26, 147-167.

Bruns, W. J. Jr., and K. A. Merchant: August 1990, The Dangerous Morality of Managing Earnings,
Management Accounting 72 (2), 22-25.

Cohen, D., and L. Prusak; 2001, In Good Company: How Social Capital Makes Organizations Work.
(Harvard Business School Press, Boston).

Crant, J. M., and T. S. Bateman: 1993, Assignment of Credit and Blame for Performance Outcomes,
Academy of Management Journal 36 (1), 7-27.

Davidson, W.N. III., Jiraporn, P., Kim, Y.S., and Nemec, C.: 2004, Earnings Management Following
Duality-Creating Successions: Ethnostatistics, Impression Management, and Agency Theory,
Academy of Management Journal 47 (2), 267-275.

Elias, R. Z.: 2004, The Impact of Corporate Ethical Values on Perceptions of Earnings Management,
Managerial Auditing Journal 19 (1), 84-98.

Elig, T. W., and I. H. Frieze: 1979, Measuring Causal Attributions for Success and Failure, Journal of
Personality and Social Psychology 37, 621-634.

Erickson, D. J., and D. S. Krull: 1999, Distinguishing Judgments About What From Judgments About
Why: Effects of Behavior Extremity on Correspondent Inferences and Causal Attributions,
Basic and Applied Social Psychology 21(1), 1-11.

Flory, S. M., T. J. Phillips, R. E. Reidenbach, and D. P. Robin: 1992, A Multidimensional Analysis of


Selected Ethical Issues in Accounting, The Accounting Review 67, 284.

Forte, A.: 2004, Business Ethics: A Study of the Moral Reasoning of Selected Business Managers and
the Influence of Organizational Ethical Climate, Journal of Business Ethics 51 , 167-173.

Green, S. G., and T. R. Mitchell: 1979, Attributional Processes of Leaders in Leader-Member


Interactions, Organizational Behavior and Human Performance 23, 429-458.

26
Guidry, F., A. J. Leone, and S. Rock: 1999, Earnings-Based Bonus Plans and Earnings Management by
Business-Unit Managers, Journal of Accounting and Economics 26, 113-142.

Hamilton, V. L.: 1980, Intuitive Psychologist or Intuitive Lawyer? Alternative Models of the
Attribution Process, Journal of Personality and Social Psychology 39(5), 767-772.

Healy, P. M.: 1985, The Effect of Bonus Schemes on Accounting Decisions, Journal of Accounting
and Economics 7, 85-107.

Healy, P. M., and J. M. Wahlen: 1999, A Review of the Earnings Management Literature and its
Implications for Standard Setting, Accounting Horizons 13(4), 365-383.

Heider, F.: 1958, The Psychology of Interpersonal Relations (John Wiley and Sons, New York, N.Y.).

Hollinger, R.C. and Clark, J.P.: 1983, Theft by Employees (Lexington Books, Lexington, MA).

Hopwood, A. G.: 1972, An Empirical Study of the Role of Accounting Data in Performance
Evaluation, Journal of Accounting Research 10, S156-S182.

Jensen, M. C.: January 8, 2001, Why Pay People to Lie? Wall Street Journal, A32.

Jones, E.E. and Nisbett, R.E.: 1972, The Actor and the Observer: Divergent Perceptions of the Causes
of Behavior. In E.E. Jones, D.E. Kanouse, H.H. Kelley, R.E. Nisbett, S. Valins, and B. Weiner
(Eds.) Perceiving the Causes of Behavior (General Learning Press, Morristown, N.J.).

Kaplan, S. E.: 2001a, Ethically Related Judgments by Observers of Earnings Management, Journal
of Business Ethics 32, 285-298.

Kaplan, S. E.: 2001b, Further Evidence on the Ethics of Managing Earnings: An Examination of the
Ethically Related Judgments of Shareholders and Non-Shareholders, Journal of Accounting
and Public Policy 20, 27-44.

Kaplan, S. E., and P. M. J. Reckers: 1985, An Examination of Auditor Performance Evaluation, The
Accounting Review 60, 477-487.

Kaplan, S. E, and P. M. J. Reckers: 1991, An Attributional Analysis of the Performance Evaluation


Process, Advances in Accounting 9, 227-248.

Kaplan, S. E., and P. M. J. Reckers: 1993, An Examination of the Effects of Accountability Tactics on
Performance Evaluation Judgments in Pubic Accounting, Behavioral Research in Accounting 5,
101-123.

Kelley H. H.: 1972, Causal Schemata and the Attribution Process, In E.E. Jones, D.E. Kanouse, H.H.
Kelley, R.E. Nisbett, S. Valins, and B. Weiner (Eds.) Perceiving the Causes of Behavior
(General Learning Press, Morristown, N.J.).

27
Kelley, H. H., and J. L. Michela: 1980, Attribution Theory and Research, Annual Review of
Psychology 31, 457-501.

Kelly, J. E.: 2004, Solidarity and Subsidiarity: Organizing Principles for Corporate Moral Leadership
in the New Global Economy, Journal of Business Ethics 52, 283-295.

Kerr, S.: 1995, On the Folly of Rewarding A While Hoping for B, Academy of Management
Executive 9, 7-14.

Kida, T.: 1984, The Impact of Hypothesis-Testing Strategies on Auditors Use of Judgment Data,
Journal of Accounting Research 22, 332-340.

Lander, G.P.: 2004, What is Sarbanes-Oxley (McGraw-Hill, New York, NY).

Lowe, D. J., P. M. J. Reckers, and S. M. Whitecotton: 2002, The Effects of Decision-Aid Use and
Reliability on Jurors Evaluations of Auditor Liability, The Accounting Review 77, 185-202.

Lord, R. G. and J. E. Smith: 1983, Theoretical, Information Processing, and Situational Factors
Affecting Attribution Theory Models of Organizational Behavior, Academy of Management
Review 8(1), 50-60.

Merchant, K. A.: 1998, Modern Management Control Systems (Prentice Hall, Upper Saddle River, NJ).

Merchant, K. A., and J. Rockness: 1994, The Ethics of Managing Earnings: An Empirical
Investigation, Journal of Accounting and Public Policy 13, 79-94.

Mowday, R. T.: 1983, Beliefs About the Causes of Behavior: The Motivational Implications of
Attributional Processes, In R. M. Steers and L. W. Porter (eds) Motivation and Work Behavior
(McGraw-Hill Book Company, New York, N.Y.), 352-374.

Noreen, E.: 1988, The Economics of Ethics: A New Perspective on Agency Theory, Accounting,
Organizations and Society 13, 359-369.

Reckers, P. M. J., and B. Wong-on-Wing: 1991, Managements Motive and its Effect on Selected Audit
Decisions, Behavioral Research in Accounting 3, 77-94.

Reeder, G. D., and J. M. Spores: 1983, The Attribution of Morality, Journal of Personality and Social
Psychology 44, 736-745.

Regan, D.T. and Totten, J.: 1975, Empathy and Attribution: Turning Observers into Actors, Journal
of Personality and Social Psychology 32, 850-856.

Sanderson, C. A., and J. M. Darley: 2002, I am Moral, but You are Deterred: Differential
Attributions about Why People Obey the Law, Journal of Applied Social Psychology 32, 375-
405.

28
Singer, M. S., and A. E. Singer: 1997, Observer Judgments About Moral Agents Ethical Decisions:
The Role of Scope of Justice and Moral Intensity, Journal of Business Ethics 16, 473-484.

Sprinkle, G. B.: 2003, Perspectives on Experimental Research in Managerial Accounting, Accounting,


Organizations and Society 28, 287-318.

Tan, H-T, and M. G. Lipe: 1997, Outcome Effects: The Impact of Decision Process and Outcome
Controllability, Journal of Behavioral Decision Making 10, 315-325.

Trevino, L.K.: 1992, The Social Effects of Punishment in Organizations: A Social Justice Perspective,
Academy of Management Review 17, 647-676.

Trotman, K. T., and J. Sng: 1989, The Effect of Hypothesis Framing, Prior Expectations and Cue
Diagnosticity on Auditors Information Choice, Accounting, Organizations and Society 14, 565-
576.

Van Der Stede, W. A.: 2000, The Relationship Between Two Consequences of Budgetary Controls:
Budgetary Slack Creation and Managerial Short-Term Orientation, Accounting, Organizations
and Society 25 (6), 609-622.

Weiner, B.: 1985a, Spontaneous Causal Thinking, Psychological Bulletin 97 (1), 74-84.

Weiner, B.: 1985b, Human Motivation. (New York: Springer-Verlag).

Wong-on-Wing, B., J. H. Reneau, and S. G. West: 1989. Auditors perception of management:


Determinants and consequences. Accounting, Organizations and Society 14: 577-587.

29
Table 1

Dependent Variable Descriptive Statistics and Correlation Matrix


(N=159)

Mean S.D. (1) (2) (3) (4) (5) (6)


____________________________________________________________________________
Internal Attributions (1) .48 3.33 -

Morality (2) 5.56 2.24 .34** -

Hardworking (3) 5.28 1.56 .06 .43** -

Knowledgeable (4) 5.52 1.57 -.04 .35** .61** -

Decision Making Ability (5) 5.62 1.91 .08 .59** .41** .50** -

Willingness to Work With (6) 5.69 2.16 .23** .75** .31** .38** .64** -

____________________________________________________________________________
**p<.01

30
Table 2

The Effect of Earnings Management, Budgetary Control Style and Work History on
Causal Attributions and Morality Assessments

Net Morality
Source Attributions Assessments
______________________________________________________________________________

F-Score F-Score

Earnings Management (EM) 20.54 *** 251.95***


(.12)a (.63)

Budgetary Control Style (BCS) 3.46 3.12

Work History (WH) 4.65* 1.74


(.03)

EM X BCS 5.44* 3.05


(.04)

EM X WH 1.65 3.85*
(.03)
BCS X WH 1.35 .96

EM X BCS X WH .08 2.30


_____________________________________________________________________________

*p<.05
**p<.01
***p<.001
a
eta 2 values presented in parentheses

31
Table 3

Descriptive Statistics Net Attributions and Morality Assessments

Panel A: Earnings management Behavior and Budgetary Control Style Means for Net Attributions
(standard deviation)

Earning Management Behavior


Did not Manage Did Manage
BCS1 rigid -0.7 (2.9) 2.6 (3.6) 1.0 (3.6)
BCS2 - flexible -0.5 (2.7) 0.5 (3.1) 0.1 (2.9)
-0.6 (2.8) 1.5 (3.5)

Panel B: Work history Means for Net Attributions (standard deviation)

Work history
Average 1.0 (3.1)
Above average 0.0 (3.5)

Panel C: Earnings management Behavior and work history means for morality
assessments (standard deviation)

Earning Management Behavior


Did not Manage Did Manage
Work history average 3.5 (1.1) 7.4 (1.7) 5.3 (2.4)
Work history above 4.2 (1.0) 7.2 (1.7) 5.8 (2.1)
3.8 (1.1) 7.3 (2.1)

32
Table 3

Regression Analysis on Target Managers Reputational Judgments

Decision Making Willingness to


Hardworking Knowledgeable Capability work with

Beta t-value Beta t-value Beta t-value Beta t-value

___________________________________________________________________________________________

Net
Attributions -.04 -1.23 -.08 -2.23* -.07 -1.92 -.02 -.51

Morality
Assessments .32 6.01*** .29 5.20*** .54 9.27*** .74 13.62***

Model R2 .19 .15 .36 .57

___________________________________________________________________________________________

*p<.05
**p<.01
***p<.001

33
Figure 1

Interaction of Earnings Management and


Budgetary Control Style on Net Attributions1

2.60

2.5

2.0
Net
Attribution 1.5

1.0

.5 No Earnings Management
.48
0

-.5 Earnings
-.45
-1.0 -.70
-1.5
| |

Rigid Flexible

Budgetary Control Style

1
Positive scores on the net attribution measure are indicative of internal attributions, while negative scores
indicate situational attributions.

34